artpar

100x.bot is completely powered by daptin, canaster.in (work in progress, i am redoing the frontend in webgl) is also running on daptin. there are various internal company services which are running on daptin which i cannot name here.

almost any project which matures to requiring a user authn/authz (and there are always assets owned by users) i default to using daptin

I have hardened it a lot over the last 10 years, especially in terms of transaction/concurrency/security. happy to help you setting it up in your project. you can dm me ( x.com/phparth ) or leave an issue on the github repo

hey I have built daptin for this and been running all of my projects with daptin as the sole backend server for data/authn/authz.

1. oauth/account management -> daptin can serve itself as a oauth provider as well as allow "login via oauth" on 3rdparty oauth servers. you get a full managed account including forget/reset password

2. permissions -> this involves more than rate limits, like who's the owner of what and you can manage permissions at user/guest/group level

3. handling payments -> fraud monitoring is usually part of the payment processor and you are not supposed to do much about it

and these days you can point your llm agent to the wiki to get most of the stuff done without needing to read the docs yourself. you just need a cheap vps (the $5 ones on digital ocean will be more than fine).

https://github.com/daptin/daptin/wiki

https://github.com/daptin/daptin/wiki/Permissions

The key takeaway pretty much applies to the authoring of the article itself. The LLM knew what all happened but couldn't put it into a readable article.

it is a javascript interpreter in javascript, based on the ast generated by acron parser. to avoid calling "eval" or "new Function" for code executed by un-trusted sources. a sandbox

screenshots in the github readme https://github.com/artpar/canopy

Here is an implementation you might like

https://github.com/artpar/terminal-tunnel

P2P with webrtc (pion ftw) with e2ee

client side is webui so you can use on any device

ps: the default Cloudflare Worker from my account is already maxed out so you will need your own exchange (self host on your account)

I tried that over a month

except '- the directness and simplicity of JavaScript'

https://github.com/artpar/guage

But somehow the language feels so foreign. it can obviously do hello world, but I don't have a real use case

PS: the "Pure symbols only" is no longer true, most symbols have been converted to English names

and, the "days" you see there in the markdowns are "claude code sessions", not actual days

I did show hn just yesterday you don't need tailscale or any 3rd party server. Just use webrtc and it's just your mobile and laptop. end 2 end encrypted. no 3rd party dependency.

https://news.ycombinator.com/item?id=46514587

Thanks! And wow, getting a kind word from a Pion maintainer means a lot. Your library made this whole thing possible. The datachannel API is incredibly clean to work with. Appreciate you and the team's work on it.

I wrote two

jslike (acorn based parser)

https://github.com/artpar/jslike

https://www.npmjs.com/package/jslike

wang-lang ( i couldn't get ASI to work like javascript in this nearley based grammar )

https://www.npmjs.com/package/wang-lang

https://artpar.github.io/wang/playground.html

https://github.com/artpar/wang

I am actually going to use this instead of the one I created ( I could never get to correct ASI handling in that grammar)

This is exactly what I wanted and couldn't find. Ended up creating along with an interpreter (so slightly easier then walk and execute)

I want to execute dynamically generated javascript looking code in chrome extension without using eval or new function. basically eval without actually using eval.

linter would help me find and avoid usages of eval.

csp safe has a particular meaning associated it with. its not a "safe" language whatever that is. chrome webstore team is okay with it and serves my purpose. if you have submitted extensions to google chrome then you would know that any sign of "eval" or new Function in the code will lead to rejection.

I made a language for using in another project, so I'll answer your questions

https://www.npmjs.com/package/wang-lang

- this new language looks and behaves exactly like javascript, except it doesnt have "eval" and "new Function", so it is CSP safe. That's the only difference. I wanted to execute dynamically generated code in chrome extension

- llm did most of the work of creating a nearley grammar and associated interpreter (whole thing is bundled, nearley is not a final dependency), elaborate tests make this quite sane to handle

- took me about total of 1 weeks for the initial mvp to try out, and then have been fixing bugs and inconsistencies with javascript behavior, about 1 day a month of effort

- mostly 0

The only reason to create was I couldnt find something similar and it was low effort thanks to llm

I also created another even smaller DSL you can say

https://www.npmjs.com/package/free-text-json-parser

It parses json embedded in plain text

100x.bot primarily a browser automation engine (think imacros ) but with llm and all the tools for interacting with the Dom and a better interface . there is a workflow builder so you do not need to rely on llm for executing deterministic workflows.

I did not know gstreamer wasm also exists, I'll check it out

I think that goes with almost every tool you want to use with llm. User should already know the tool ideally so mistakes by llm can be prevented before they happen.

Here making ffmpeg as "just another capability" allows it to be stitched together in workflows

I am sorry I think you might have installed the our workflow plugin(close ended agent) instead of agent4 (open ended). Its a new experiment we are working. We are working on the website messaging.

Agent4 - https://chromewebstore.google.com/detail/agent4/kipkglfnhnpb...

Old workflow plugin - https://chromewebstore.google.com/detail/100xbot/dhcenlmiiom...

No it is not open-source. But it is not obfuscated either, so you can always look into the code by downloading the plugin from chrome webstore if (and these days llms can help with that a lot) if you are into that kind of verification.

That's exactly what is it doing. The workflows are pretty much js snippets in itself you can see in the "code" tab (in the plugin when you select a saved workflow).

Yeah like so many legacy things, unfortunately they are not going away that fast. People are still clicking on these tedious interfaces day in day out to get all the "smaller" stuff running. Even if every one agreed on the "One Best UI", it would take decades to convert all the existing ones before breaking a lot of flows.

Ask the agent to make workflows for you.

You can use floating window or sidepanel also for the agent. New tab is just for convenience.

Worst case scenario we can just shut down sharing/public workflows altogether, or do you have something else in mind ?

One shot is pretty much not going to work, both at single step level or if you ask llm to generate workflow in one shot. We haven't measured it as such but even for static websites like hackernews front page it takes a couple tries of to and fro for the llm to get it right. somehow after all the instructions the llm will still "guess" the selector instead of checking the page/dom contents. And then there are lot of other minor details that need to be captured like "you need to wait a couple of second for the auto complete results to show up". If you tell it to just make a workflow, it will generate some garbage and call it a day.

I don't know if plywright works without chrome in debug mode, but I tried the MCP for chrome devtools and it requires chrome to be started in debugging mode and that basically means you cant log into a lot of sites (especially google) since it will block you with an "Unsafe" message. Works pretty well if you owe the target website.

So actually I have this setup (of a bridge server) which I use for agent4 itself (so claude code can talk to agent4), It makes a lot of sense to publish that bridge as well in the MCP form.

Yes I can imagine PI somehow being stored in the workflow. I frequently see llms hardcoding tests just to make user happy and this can also happen in the browser version where if something is too hard to scrape but agent is able to infer from screenshot so it might end up making a workflow that seems correct but is just hardcoded with data. We are thinking of multiple guards/blocks to not let user create such a workflow, but the risks that come with an open ended agent are still going to be present.

If I am reading you correctly, you captured the whole essence of agent4.

So it does the first pass (based on your goals) makes memories (and these are local)

Now you tell the agent you want to do this repeatedly, so it will make a workflow (these workflows are saved on server, currently all public for now but we are working out permissions/group based access) for you based on these memories and interactions.

The problem is many times that the agent thinks is stable isn't really, so there a feedback loop for the agent to test out the workflow and improve them. (its basically claude code/codex sitting in the browser)

Workflow details are appended to prompt based on user query match/opened tabs match.

Everyone thinks of typical e-commerce pages when its comes "browser agent doing something", but our real use cases are far from shopping for the user. But your point still stands valid. The idea is that maybe there are websites where generating stable selectors/hierarchy maps wouldn't solve, but 80% (from 80-20) of websites are not like that (including a lot of internal dashboard/interfaces) (there will also be issues for websites with proper i18n implementations if the selectors are aria label based)

Self healing css selectors is also only 1 part of the story. The other part is the cohesive interface for the agent itself to use these selectors.